CN108881453A - Communication means and its network node under a kind of distributed file system - Google Patents
Communication means and its network node under a kind of distributed file system Download PDFInfo
- Publication number
- CN108881453A CN108881453A CN201810684790.8A CN201810684790A CN108881453A CN 108881453 A CN108881453 A CN 108881453A CN 201810684790 A CN201810684790 A CN 201810684790A CN 108881453 A CN108881453 A CN 108881453A
- Authority
- CN
- China
- Prior art keywords
- network node
- client
- file system
- cluster
- target network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/14—Session management
- H04L67/141—Setup of application sessions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/14—Session management
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
This application provides the communication means under a kind of distributed file system, the method is applied to distributed file system, and the cluster of the distributed file system includes multiple network nodes, the method includes:When the target network node in the multiple network node will rejoin in the cluster, determine that the target network node enters the pre-connection stage;When the target network node is in the pre-connection stage, the survival client in the target network node and the distributed file system carries out network reconnection respectively, and the target network node is without business retransmission process;After the pre-connection stage is overtime, determining that the target network node enters access phase, the client of the target network node and successful connection executes business respectively to be retransmitted and operates, wherein, after the target network node enters access phase, the network reconnection operation with client is not executed.
Description
Technical field
This application involves the communications fields, and more particularly, under a kind of distributed file system communication means and
Its network node.
Background technique
In distributed file system, network node in the cluster delay machine or to restart be very universal due to a variety of causes
Event, network node can generally rejoin automatically cluster after fault recovery.Restart in network node and reverts to collection again
During group members, generally thering is a stage to be client reconnection --- this node can wait old client all complete with it
At reconnection and business recovery, or wait some long period after because of this stage of overtime interrupt.But current mainstream
In distributed file system, the waiting time in this stage causes the waiting time long, leads because a variety of causes and considering
Before cause the network node of delay machine rejoin cluster speed it is slow, and then when having tied down the business recovery of entire cluster
It is long.
Therefore, the communication means under a kind of distributed file system is needed, network node can be shortened and access cluster again
Communication speed.
Summary of the invention
The application provides the communication means under a kind of distributed file system, can shorten network node and access cluster again
Communication speed.
In a first aspect, the communication means under providing a kind of distributed file system, the method is applied to distributed text
Part system, the cluster of the distributed file system include multiple network nodes, the method includes:When the multiple network section
When target network node in point will be rejoined in the cluster, determine that the target network node enters pre-connection rank
Section;When the target network node is in the pre-connection stage, the target network node and the distributed field system
Survival client in system carries out network reconnection respectively, and the target network node is without business retransmission process;Work as institute
After stating pre-connection stage time-out, determine that the target network node enters access phase, the target network node with connected
It connects successful client and executes business repeating transmission operation respectively, wherein after the target network node enters access phase, do not execute
With the network reconnection operation of client.
With reference to first aspect, in the first possible implementation of the first aspect, the method also includes:When described
Destination network device detect in the distributed file system initiate network reconnection requests the first client and with it is described
After first client successful connection, determine first client for survival client, and will first client addition etc.
To in queue.
With reference to first aspect and its above-mentioned implementation, in the second possible implementation of the first aspect, described
Target network node and the client of successful connection execute business respectively and retransmit operation, including:The target network node
Business successively, which is carried out, with the client in the waiting list retransmits operation.
With reference to first aspect and its above-mentioned implementation, in a third possible implementation of the first aspect, described
Target network node is to rejoin the network node of the cluster after fault restoration.
With reference to first aspect and its above-mentioned implementation, in a fourth possible implementation of the first aspect, described
The pre-connection stage, duration was by using the client terminal quantity of the cluster to determine.
Second aspect, provides a kind of network node, and the network node applies are described in the cluster of distributed file system
Cluster includes multiple network nodes, and the network node includes:Determination unit, the determination unit are used to work as the network node
When will rejoin in the cluster, determine that the network node enters the pre-connection stage;Processing unit, the processing unit
For being distinguished with the survival client in the distributed file system when the network node is in the pre-connection stage
Network reconnection is carried out, and without business retransmission process;The processing unit is also used to after the pre-connection stage is overtime,
It determines that the network node enters access phase, executes business respectively with the client of successful connection and retransmit operation, wherein
After the network node enters access phase, the network reconnection operation with client is not executed.
In conjunction with second aspect, in the first possible implementation of the second aspect, the processing unit is also used to:When
The network equipment detect in the distributed file system initiate network reconnection requests the first client and with it is described
After first client successful connection, determine first client for survival client, and will first client addition etc.
To in queue.
It is in a second possible implementation of the second aspect, described in conjunction with second aspect and its above-mentioned implementation
Processing unit is used for:Business successively, which is carried out, with the client in the waiting list retransmits operation.
It is in the third possible implementation of the second aspect, described in conjunction with second aspect and its above-mentioned implementation
The pre-connection stage, duration was by using the client terminal quantity of the cluster to determine.
Based on the above issues, the present invention proposes that one kind in file system nodes recovery process, shortens client reconnection rank
The time-consuming mechanism of section, so that the bonding ratio of this stage old client can not be influenced and substantially to shorten old client complete
At the time of reconnection, and then the speed of group service recovery after node delay machine is substantially improved, shortens fail-over time.
Detailed description of the invention
Fig. 1 is the schematic flow chart of the method for the application one embodiment.
Fig. 2 is the schematic block diagram of the network node of the application one embodiment.
Specific embodiment
Below in conjunction with attached drawing, the technical solution in the application is described.
Firstly, following concept is introduced, the network node of distributed file system:One computer node, thereon
Distributed file system is installed.
Distributed file system cluster:The node set that one group of distribution file system nodes is constituted, it is common externally to provide
File system service.
Cluster:The abbreviation of distributed file system cluster.
Client:Computer node used in client provides a component thereon, to access the file of cluster offer
System service.
Old client:The client of service connection is established with cluster before.
As shown in Figure 1, showing the schematic flow chart of the method for the application one embodiment, the method is applied to divide
Cloth file system, the cluster of the distributed file system include multiple network nodes, including:
Step 110, when the target network node in the multiple network node will rejoin in the cluster, really
The fixed target network node enters the pre-connection stage;
Step 120, when the target network node is in the pre-connection stage, the target network node with it is described
Survival client in distributed file system carries out network reconnection respectively, and the target network node is without business weight
Hair processing;
Step 130, after the pre-connection stage is overtime, determine that the target network node enters access phase, it is described
Target network node and the client of successful connection execute business respectively and retransmit operation, wherein the target network node
Into after access phase, the network reconnection operation with client is not executed.
Network layer reconnection of the current main-stream system the reconnection stage and operation layer are retransmitted mixed in together, and the problem of bringing is
Because node also has the internal logic of other complexity to need to handle at the moment, often have no idea to immediately treat each old client
The reconnection requests at end in a disguised form cause reconnection requests to be delayed by processing, thus node must not be not provided with a long waiting time,
The connection request of all old clients is completed with this.
Therefore, the process that destination node is accessed to cluster again is split as the purpose in two stages, is by pre-connection
Stage waits merely the network reconnection of old client, and access phase handle merely client business retransmit or other inside
Request, there is no other complex logics to interfere the pre-connection stage in this way, so as to which shorter surpass is arranged to the pre-connection stage
When the time.
Optionally, as the application one embodiment, the method also includes:When the destination network device detects institute
State in distributed file system initiate network reconnection requests the first client and with after the first client successful connection,
It determines that first client is survival client, and first client is added in waiting list.
Specifically, the pre-connection stage quickly determines which old client is also survived, and reject non-viable client.?
The pre-connection stage does not do other time-consuming business processings, waits the reconnection of old client merely.
If detecting the network reconnection requests of client, then it is assumed that old client is also survived, and survival is added into
List;If the pre-connection stage is overtime, then it is assumed that there are no the old clients that connection comes up, and failure, is no longer waiting for it,
And then it is transferred to access phase;The pre-connection period is generally arranged shorter, such as can be 1-3 seconds, because multiple without other
The interference of miscellaneous logic can complete the processing of the network reconnection requests of old client in the very short period.
Optionally, as the application one embodiment, the pre-connection stage duration is by using the client of the cluster
Quantity determines.
Optional volume, as one embodiment in this community, the client of the target network node and successful connection
Business is executed respectively retransmits operation, including:The target network node successively carries out industry with the client in the waiting list
Operation is retransmitted in business.
Since in the pre-connection stage, the request that the business of old client retransmits request or intra-node is put into
Waiting list so this stage is needed again to take out from waiting list, and executes them.Specifically, waiting list before taking-up
In request, sequentially execute them, after the completion of execution, access phase complete.
Optionally, as the application one embodiment, the target network node is described to rejoin after fault restoration
The network node of cluster.
Therefore, the present invention rejoins collection after some node failure by improving in distributed file system cluster
When group, the mechanism of old client reconnection is waited, while guaranteeing that old client can be connected to node as far as possible, is substantially contracted
The time-consuming in short client reconnection stage, so accelerate clustered node resume speed, guarantee entire file system availability and
Experience property.
Fig. 2 shows the schematic block diagrams of the device of the network node of the application one embodiment, as shown in Fig. 2,
The network node 200 is applied to the cluster of distributed file system, and the cluster includes multiple network nodes, the network section
Putting 200 includes:Determination unit 210, the determination unit 210 are used to that the cluster will to be rejoined when the network node 200
When in the middle, determine that the network node 200 enters the pre-connection stage;Processing unit 220, the processing unit 220 is for working as institute
When stating network node 200 and being in the pre-connection stage, carried out respectively with the survival client in the distributed file system
Network reconnection, and without business retransmission process;The processing unit 220 is also used to after the pre-connection stage is overtime,
It determines that the network node 200 enters access phase, executes business respectively with the client of successful connection and retransmit operation,
In, after the network node 200 enters access phase, the network reconnection operation with client is not executed.
Optionally, as the application one embodiment, the processing unit 200 is also used to:When the network equipment 200
It detects the first client for initiating network reconnection requests in the distributed file system and connects with first client
After being connected into function, determine that first client is survival client, and first client is added in waiting list.
Optionally, as the application one embodiment, the processing unit 200 is also used to:Successively with the waiting list
In client carry out business retransmit operation.
Optionally, as the application one embodiment, the processing unit 200 is also used to:The pre-connection stage duration
By using the client terminal quantity of the cluster to determine.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, it can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, the indirect coupling of device or unit
It closes or communicates to connect, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or the second equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes:USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (9)
1. the communication means under a kind of distributed file system, which is characterized in that the method is applied to distributed file system,
The cluster of the distributed file system includes multiple network nodes, the method includes:
When the target network node in the multiple network node will rejoin in the cluster, the target network is determined
Network node enters the pre-connection stage;
When the target network node is in the pre-connection stage, the target network node and the distributed field system
Survival client in system carries out network reconnection respectively, and the target network node is without business retransmission process;
After the pre-connection stage is overtime, determine that the target network node enters access phase, the target network node
Business is executed respectively with the client of successful connection and retransmits operation, wherein the target network node enters access phase
Afterwards, the network reconnection operation with client is not executed.
2. the method according to claim 1, wherein the method also includes:
When the destination network device detects the first client of initiation network reconnection requests in the distributed file system
And after the first client successful connection, determine that first client is survival client, and objective by described first
Family end is added in waiting list.
3. according to the method described in claim 2, it is characterized in that, the client of the target network node and successful connection
End executes business respectively and retransmits operation, including:
The target network node successively carries out business with the client in the waiting list and retransmits operation.
4. according to the method described in claim 3, it is characterized in that, the target network node be fault restoration after rejoin
The network node of the cluster.
5. according to the method described in claim 4, it is characterized in that, the pre-connection stage duration is by using the visitor of the cluster
Family end quantity determines.
6. a kind of network node, which is characterized in that the network node applies are in the cluster of distributed file system, the cluster
Including multiple network nodes, the network node includes:
Determination unit, the determination unit are used for when the network node rejoins in the cluster, described in determination
Network node enters the pre-connection stage;
Processing unit, the processing unit is used for when the network node is in the pre-connection stage, with the distribution
Survival client in file system carries out network reconnection respectively, and without business retransmission process;
The processing unit is also used to determine that the network node enters access phase after pre-connection stage is overtime, with
The client of successful connection executes business respectively and retransmits operation, wherein after the network node enters access phase, does not hold
The capable network reconnection operation with client.
7. network node according to claim 6, which is characterized in that the processing unit is also used to:
When the network equipment detect in the distributed file system initiate network reconnection requests the first client and
After the first client successful connection, first client is determined for survival client, and by first client
It is added in waiting list.
8. network node according to claim 7, which is characterized in that the processing unit is used for:
Business successively, which is carried out, with the client in the waiting list retransmits operation.
9. network node according to claim 8, which is characterized in that the pre-connection stage duration is by using the cluster
Client terminal quantity determine.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810684790.8A CN108881453A (en) | 2018-06-28 | 2018-06-28 | Communication means and its network node under a kind of distributed file system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810684790.8A CN108881453A (en) | 2018-06-28 | 2018-06-28 | Communication means and its network node under a kind of distributed file system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108881453A true CN108881453A (en) | 2018-11-23 |
Family
ID=64296170
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810684790.8A Pending CN108881453A (en) | 2018-06-28 | 2018-06-28 | Communication means and its network node under a kind of distributed file system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108881453A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2005125056A1 (en) * | 2004-06-22 | 2005-12-29 | Zte Corporation | A method for service connection setup and service resume protection in optical network |
CN101102521A (en) * | 2006-07-03 | 2008-01-09 | 华为技术有限公司 | A method for service protection in automatically switching optical network |
CN101494801A (en) * | 2008-01-21 | 2009-07-29 | 华为技术有限公司 | Method, system and network appliance for fault recovery |
CN101682518A (en) * | 2007-06-01 | 2010-03-24 | 北方电讯网络有限公司 | Distributed connection establishment and restoration |
CN102487329A (en) * | 2010-12-02 | 2012-06-06 | 中兴通讯股份有限公司 | Service restoring method and device |
CN103581225A (en) * | 2012-07-25 | 2014-02-12 | 中国银联股份有限公司 | Distributed system node processing task method |
-
2018
- 2018-06-28 CN CN201810684790.8A patent/CN108881453A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2005125056A1 (en) * | 2004-06-22 | 2005-12-29 | Zte Corporation | A method for service connection setup and service resume protection in optical network |
CN101102521A (en) * | 2006-07-03 | 2008-01-09 | 华为技术有限公司 | A method for service protection in automatically switching optical network |
CN101682518A (en) * | 2007-06-01 | 2010-03-24 | 北方电讯网络有限公司 | Distributed connection establishment and restoration |
CN101494801A (en) * | 2008-01-21 | 2009-07-29 | 华为技术有限公司 | Method, system and network appliance for fault recovery |
CN102487329A (en) * | 2010-12-02 | 2012-06-06 | 中兴通讯股份有限公司 | Service restoring method and device |
CN103581225A (en) * | 2012-07-25 | 2014-02-12 | 中国银联股份有限公司 | Distributed system node processing task method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3720050A1 (en) | Method and apparatus for deploying network slice | |
US8788565B2 (en) | Dynamic and distributed queueing and processing system | |
CN111641515B (en) | VNF life cycle management method and device | |
EP3182678A1 (en) | Method for upgrading network function virtualization application, and method and device for forwarding service | |
WO2012132815A1 (en) | Virtual server id management system, integrated monitoring system, virtual server id management program, and integrated monitoring program | |
CN105978781A (en) | Method and system for establishing network connection of Docker container, and client side | |
CN107172187B (en) | A kind of SiteServer LBS and method | |
CN111866064A (en) | Load balancing method, device and system | |
CN109375933B (en) | Method and equipment for upgrading client | |
WO2016197346A1 (en) | Method and apparatus for managing life circle of virtualized network function | |
CN112019366B (en) | Leasing method and device of physical host, cloud platform and readable storage medium | |
CN109358967A (en) | A kind of ME platform APP instantiation moving method and server | |
CN111858083B (en) | Remote service calling method and device, electronic equipment and storage medium | |
CN107688502A (en) | Inter-process communication methods and device | |
CN104793981B (en) | A kind of online snapshot management method and device of cluster virtual machine | |
CN109491895A (en) | Server stress test method and device | |
CN109845192A (en) | Computer system and method for the network that dynamically internet adapter software defines | |
CN110365517A (en) | Data processing method, device, storage medium and computer equipment | |
CN111090519A (en) | Task execution method and device, storage medium and electronic equipment | |
CN107548038A (en) | Method for processing short messages and device | |
CN108667750B (en) | Virtual resource management method and device | |
CN109976883A (en) | A kind of processing method and its system of task | |
CN111355602B (en) | Resource object management method and device | |
CN108881453A (en) | Communication means and its network node under a kind of distributed file system | |
CN116886286A (en) | Big data authentication service self-adaption method, device and equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181123 |
|
RJ01 | Rejection of invention patent application after publication |